Unpleasant it may be, but investors do better if the mutual funds they hold are managed by companies with an ethos more like that of a viper pit than a kindergarten.
Cooperation is essential for wellbeing and economic growth, surely, but a new study shows that fund families with more internal competition for rewards among managers produce better results.
The study, covering a range of mutual fund equity sectors between 1992 and 2015, looked at underlying compensation mechanisms as well as other indicators to gain insight into which fund families had relatively more cooperative or competitive environments.
The upshot is that competition among fund managers, featuring a greater relationship between an individual's performance and her pay, tended to produce better results for investors.
"We find that competitive incentives result in higher average performance consistent with either incentivizing greater effort on the part of managers or attracting managers with higher ability," Richard Evans of the University of Virginia and Melissa Porras Prado and Rafael Zambrana of the Nova School of Business and Economics write in a working paper released in July.
The study constructed a series of indices seeking to measure how much managers in a given fund family cooperate or compete. These indices measured, for example, how managers are compensated as well as how likely they were to work as part of a team and how large those teams were.
This builds on 2013 research that showed that managers can cooperate with each other if incentivized. That cooperative behavior, which presumably is driven in substantial part by compensation rules, can result in things like funds at a given family buying shares from sibling funds which are facing high redemptions.
That can be good for the fund firm, which is able to protect its investment in the brand and wellbeing of a given fund, but not so great for its investors, whose best interests are unlikely to be served by asset allocation decisions calculated to help the firm.
Alternatively, fund firms in which an individual's results led more directly to her compensation tended to show less of this behavior.
The amount of outperformance among more competitive fund families is significant. The average fund in a family deemed "competitive" by the study's standards outperforms by 30 basis points per month, well over 3 percentage points a year, as compared to the median returns of equivalent funds of other families.
That's a huge annual outperformance, given that a typical mutual fund is unlikely to average more than 10 per cent a year.
Unsurprisingly, the more cooperative fund families tend to serve less sophisticated individual investors, reaching them through a broker-based distribution channel, while the more competitive firms have a bigger weighting of institutions among their clients. While the study does not demonstrate this, it is tempting to speculate that fund families which serve less savvy small investors prioritize their own wellbeing to a greater extent, simply because they can get away with it more easily. Remember, cooperation in this context is leading to behavior which benefits the firm, but with costs to the clients.
For every one standard deviation increase in competitive incentives, the practice of cross-trading – one fund in a family buying shares from another – drops by 0.18 percentage point, which is an increase compared to the average fund of a meaningful 10 per cent.
Cross-holdings, when funds hold the same shares as their in-family peers, also decrease when fund families are more competitive, the study showed.
"If the objective is to maximize the overall value of the investment advisor, coordinated action among fund managers to cross-subsidize funds of high value to the advisor at the expense of low-value funds is an important tool to accomplish this objective," the study said.
Cooperative fund firms are also more likely to capture outflows from their funds into other funds in the family, perhaps because of the role of the brokerage intermediary.
To be sure, the benefits of competition are not unalloyed or absolute. Earlier studies have found that more explicit manager performance incentives produce higher performance but significantly higher risk taking as well. It is also true that more cooperative fund families show a tighter dispersion in their funds' returns, which might be an indicator of lower risk.
While the economist's conceit that people are profit-maximizing often doesn't hold true in society, in financial services it seems pretty close to the mark.
If investors are going to use active management they should make this work for them.